Changes in Capsular Serotype Alter the Surface Exposure of Pneumococcal Adhesins and Impact...

Changes in Capsular Serotype Alter the Surface Exposureof Pneumococcal Adhesins and Impact VirulenceCarlos J. Sanchez1., Cecilia A. Hinojosa1., Pooja Shivshankar1, Catherine Hyams2, Emilie Camberlein2,

Jeremy S. Brown2, Carlos J. Orihuela1*

1 Department of Microbiology and Immunology, The University of Texas Health Science Center, San Antonio, Texas, United States of America, 2 Department of Medicine,

Centre for Respiratory Research, Royal Free and University College Medical School, Rayne Institute, London, United Kingdom

Abstract

We examined the contribution of serotype on Streptococcus pneumoniae adhesion and virulence during respiratory tractinfection using a panel of isogenic TIGR4 (serotype 4) mutants expressing the capsule types 6A (+6A), 7F (+7F) and 23F(+23F) as well as a deleted and restored serotype 4 (+4) control strain. Immunoblots, bacterial capture assays withimmobilized antibody, and measurement of mean fluorescent intensity by flow cytometry following incubation of bacteriawith antibody, all determined that the surface accessibility, but not total protein levels, of the virulence determinantsPneumococcal surface protein A (PspA), Choline binding protein A (CbpA), and Pneumococcal serine-rich repeat protein(PsrP) changed with serotype. In vitro, bacterial adhesion to Detroit 562 pharyngeal or A549 lung epithelial cells wasmodestly but significantly altered for +6A, +7F and +23F. In a mouse model of nasopharyngeal colonization, the number of+6A, +7F, and +23F pneumococci in the nasopharynx was reduced 10 to 100-fold versus +4; notably, only mice challengedwith +4 developed bacteremia. Intratracheal challenge of mice confirmed that capsule switch strains were highly attenuatedfor virulence. Compared to +4, the +6A, +7F, and +23F strains were rapidly cleared from the lungs and were not detected inthe blood. In mice challenged intraperitoneally, a marked reduction in bacterial blood titers was observed for thosechallenged with +6A and +7F versus +4 and +23F was undetectable. These findings show that serotype impacts theaccessibility of surface adhesins and, in particular, affects virulence within the respiratory tract. They highlight the complexinterplay between capsule and protein virulence determinants.

Citation: Sanchez CJ, Hinojosa CA, Shivshankar P, Hyams C, Camberlein E, et al. (2011) Changes in Capsular Serotype Alter the Surface Exposure of PneumococcalAdhesins and Impact Virulence. PLoS ONE 6(10): e26587. doi:10.1371/journal.pone.0026587

Editor: Eliane Namie Miyaji, Instituto Butantan, Brazil

Received June 2, 2011; Accepted September 29, 2011; Published October 19, 2011

Copyright: � 2011 Sanchez et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: Dr. Sanchez is supported through the United States National Institutes of Health, National Institute of Dental and Craniofacial Research grant DE14318for the Craniofacial Oral-Biology Student Training in Academic Research (COSTAR) program. Dr. Hinojosa was supported by the Astor Foundation and Glaxo SmithKline through the University College London, MB, PhD program. Dr. Camberlein was supported by the United Kingdom Medical Research Council. Dr. Orihuela issupported by United States National Institutes of Health, National Institute of Allergy and Infectious Diseases grant AI078972. For work at University CollegeLondon Hospitals/University College London a proportion of funding was obtained from the United Kingdom Department of Health, National Institute for HealthResearch Biomedical Research Centre’s funding scheme. The University of Texas Health Science Center Flow Cytometry Core Facility is supported by U.S. NationalInstitutes of Health, National Cancer Institute grant CA54174 and UL1 RR025767. The funders had no role in study design, data collection and analysis, decision topublish, or preparation of the manuscript.

Competing Interests: Dr. Hinojosa was supported by an unrestricted educational grant from Glaxo Smith Kline through the University College London, MD,PhD program. This does not alter the authors9 adherence to all the PLoS ONE policies on sharing data and materials. All other authors have declared that nocompeting interests exist.

* E-mail: [email protected]

. These authors contributed equally to this work.

Introduction

Capsular polysaccharide (capsule) is the most important

virulence determinant of Streptococcus pneumoniae, surrounding and

protecting the bacterium against complement-mediated phagocy-

tosis and bacterial killing by extracellular traps [1,2,3,4]. The

requirement for capsule is made evident by the fact that all

invasive clinical isolates of S. pneumoniae are encapsulated [5],

unencapsulated pneumococci are incapable of causing pneumonia

or sepsis in animal models [6,7], and antibodies against capsule

promote opsonization and confer protection [8,9]. Indeed,

purified capsule serves as the protective antigen in all vaccines

approved worldwide against S. pneumoniae.

Although capsule affords protection against host-clearance, it

also hinders the ability of the pneumococcus to adhere to host

epithelial cells [10,11]. This is likely due to a blocking of bacterial

adhesins by the capsule that prevents their interaction with

host ligands. To compensate, the pneumococcus spontaneously

undergoes phase variation, alternating between a transparent

low-capsule phenotype which is adhesive to host cells but subject

to opsonophagocytosis, and an opaque high-capsule phenotype

that is resistant to opsonophagocytosis but less able to attach to

cells [12]. Additionally, some pneumococci carry Pneumococcal

serine-rich repeat protein (PsrP), a member of the serine-rich

repeat protein family, or pili, that are thought to extend outward

through the capsule and mediate adhesion even in the presence of

capsule [13,14,15,16,17]. Finally, studies using scanning electron

microscopy suggest that the pneumococcus actively down-

regulates capsule production when attached to epithelial cells

[11]. Thus, the presence of capsule profoundly affects how the

pneumococcus interacts with host epithelial cells and not just

phagocytes.

PLoS ONE | www.plosone.org 1 October 2011 | Volume 6 | Issue 10 | e26587

To date more than 90 chemically and serologically and distinct

capsule types of S. pneumoniae have been described. Importantly,

not all serotypes are equally capable of colonizing the nasopharynx

or causing invasive disease. For example, in the United States only

13 serotypes (i.e. 1, 3, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19A, 19F, and

23F) are responsible for 80–90%of invasive pneumococcal

infections [18,19,20]. The reasons for why some serotypes are

more likely to cause disease remain unclear, as pneumococci are

genetically diverse and non-invasive clones belonging to these

disease-associated capsule types have also been described [21,22].

Thus the only way to assess the individual contribution of capsule

type to virulence is to compare isogenic capsule-switched strains of

the same genetic background.

In 1994, studies by Kelly et al. showed that isogenic changes in

capsule type unpredictably altered virulence following intraperito-

neal challenge. Replacement of capsule type 5 on a highly virulent

strain with type 3 resulted in a complete loss of virulence. In

contrast, change of a non-virulent serotype 6A strain to capsule type

3 enhanced virulence [23]. One proposed explanation for this was

that capsule type affected resistance to both complement deposition

and opsophagocytic uptake. In support of this notion, recent studies

have shown that the switching of a serotype 4 isolate to serotype 23F

or 6A increased C3b deposition and phagocytic uptake, whereas

replacement with 7F had a minimal effect [24]; similar results have

been reported by Melin et al. [25,26,27]. Importantly, differences in

C3b deposition on the bacterial surface have been suggested to be

due, in part, to differences in the surface exposure of pneumococcal

proteins with anti-complement activity such as Choline binding

protein A (CbpA), which binds to Factor H and is an adhesin for

laminin receptor and the polymeric immunoglobulin receptor

[28,29,30], and Pneumococcal surface protein A (PspA), which

inhibits complement deposition by unknown mechanisms [31]. For

example, replacement of strain D39 serotype 2 capsule with

serotype 3 reduced the reactivity of antisera against PspA [32].

Thus, interplay between capsule and surface exposed virulence

determinants most likely occurs and impacts virulence potential.

Herein, using isogenic mutants, we tested whether capsule type

also impacts pneumococcal adhesion and virulence at sites where

bacterial attachment is required. We determined that changes in

capsule altered surface accessibility of the pneumococcal adhesins

PsrP and CbpA as well as PspA, modulated bacterial adhesion in

vitro, and attenuated nasopharyngeal colonization and virulence

within the lungs. These studies provide insight into the complex

interplay between capsule and bacterial surface components and

help explain why not all combinations are compatible for virulence.

Methods

Bacterial strains and tissue culture cell lines usedThe isogenic capsule switch opaque phase variant mutants were

kindly provided as a gift by Dr. Jeffery Weiser at The University of

Pennsylvania and were generated using a Janus cassette [24]. S.

pneumoniae serotype 4, strain TIGR4 was used as the background

strain [21,33]. The capsule switch strains include TIGR4

derivatives expressing the 6A (+6A), 7F (+7F), and 23F (+23F)

capsule type as well as a deleted and restored capsule type 4 (+4).

The latter served as the control for the panel of isogenic mutants.

Previously, we have shown that +6A, +7F, and +23F carry

equivalent amounts of capsule as +4 and have comparable growth

rates in Todd Hewitt Broth (THB) and serum [24]. An

unencapsulated derivative of TIGR4, T4R, was also used as a

negative control [34]. Unless otherwise indicated, pneumococci

were grown on tryptic soy blood agar plates (Remel, USA) or THB

at 37uC in 5%CO2.

Confirmation of appropriate capsule production byswitch mutants

Multiplex PCR using genomic DNA as template and serotype

specific primers was performed to verify the presence of specific

capsule type genes in each strain [35,36]. Production of the correct

polysaccharide capsule was confirmed by agglutination using type-

specific antibodies (Statens Serum Institut, Denmark).

Assessment of surface accessibility of the pneumococcaladhesins on antibody- immobilized surfaces

Polystyrene 24-well plates were coated with rabbit polyclonal

antibodies against CbpA and PsrP (1:250 in PBS) overnight.

Serum from a naı̈ve animal was used as a negative control.

Bacterial cultures were diluted in phosphate buffered saline (PBS)

to an OD620 of 0.1, and incubated for 1 h at 37uC on the antibody

coated wells. Following incubation, wells were gently washed 3

times with PBS to remove unbound bacteria and attached bacteria

were freed with gentle scraping. Bacterial adhesion was deter-

mined by addition of 100 ml PBS, plating of the bacterial

suspension, and extrapolation from colony counts following

overnight incubation. Each experiment contained at least 3

biological replicates for each strain tested.

Detection of surface expression of pneumococcalproteins by flow cytometry

Indirect immunofluorescence was carried out to determine the

ability of antibodies against recombinant CbpA, PsrP and PspA

to bind to the surface of intact S. pneumoniae (17). Antibodies

against PsrP were previously generated [13,37], whereas those

against CbpA where a kind gift from Dr. Elaine Tuomanen

(Memphis, TN). Antibodies against PspA were obtained from

Santa Cruz Biotechnology (sc-17483, Santa Cruz, CA). Con-

firmed opaque pneumococci were harvested directly from blood

agar plates grown overnight or from exponential growth phase

liquid cultures (Optical Density [OD]620 = 0.5), washed in sterile

PBS and suspended in staining buffer (0.05% sodium azide and

1% bovine serum albumin in PBS). Approximately 26107

bacteria were incubated with 10% rabbit antisera to CbpA or

PsrP or with 10% goat antiserum to PspA. As a control rabbit

and goat IgG isotypes were used (Sigma Aldrich, St. Louis MO).

Following incubation at 4uC for 1 h, bacteria were washed and

incubated with a 1:100 dilution of a F(ab’)2 fragment of goat anti-

rabbit IgG (H+L) or donkey anti-goat conjugated to Phycoery-

thrin (Jackson Immunoresearch Laboratories, Westgrove PA).

Bacteria were washed in PBS and subjected to flow cytometry

using a Becton Dickinson LSR-II flow cytometer using forwards

and side scatter parameters to gate on at least 10,000 bacteria.

Results were analyzed using FlowJo v9.3 software (Becton

Dickinson) and the data is presented as mean fluorescence

intensity (MFI) and as the percent of the bacterial population

positive for relative markers.

Bacterial Adhesion AssaysA549 (human lung alveolar epithelium cell line; ATCC CRL-

185) and Detroit 562 (human nasopharyngeal epithelial cells;

ATCC CCL-138) were maintained in either F-12 media

supplemented with 10% fetal bovine serum or MEM media

supplemented with 10% FBS, 10% lactoalbumin hydrolate, and 2

mM L-glutamine, respectively. All cell lines were maintained at

37uC in 5% CO2. Adhesion assays were performed as previously

described [37]. Cells were grown to 90–95% confluence (,106

cells/well) in 24-well tissue culture polystyrene plates. Cells were

washed then exposed to media containing 107 CFU/ml of

Effect of Capsule Type on Pneumococcal Virulence


pneumococci for 1 h at 37uC in 5% CO2. Non-adherent bacteria

were removed by washing the cells 3 times with PBS. Adherent

bacteria were quantitated by lysis of the cell monolayer with 0.1%

Triton X-100 in PBS, serial dilution of the lysate followed by

plating on blood agar plates, and extrapolation from colony counts

the next day. Each experiment contained at least 3 biological

replicates for each strain tested.

Virulence studies in miceChallenge experiments described below were performed at

the University of Texas Health Science Center at San Antonio

using an Institutional Animal Care and Use Committee

approved protocol 09022-34. To minimize distress mice were

anesthetized during experimental challenge and prior to

euthanasia. Five-week-old, female, BALB/cJ mice (The Jackson

Laboratories, Bar Harbor, ME) were anesthetized with 2.5%

vaporized isoflurane. We have previously used BALB/c mice to

determine the anatomical site-specific contribution of numerous

pneumococcal virulence determinants including PsrP and CbpA

[30,37,38].

The intranasal, intratracheal, and intraperitoneal challenge

models have been previously described [37]. For intranasal

challenge, anesthetized mice were held upright and 106 CFU of

S. pneumoniae in 25 ml PBS was administered into the left nostril in

a drop-wise fashion. For intratracheal challenge, anesthetized

mice were hung upright by their incisors and 105 CFU of S.

pneumoniae in 100 mL of PBS was placed at the back of the throat.

Forced aspiration was induced by gently pulling the tongue

outward and covering the nostrils. For intraperitoneal challenge

mice were injected with 104 CFU in 100 ml PBS. On designated

days, nasopharyngeal lavage with 10 ml saline or collection of

blood from the tail vein was performed. Nasopharyngeal lavage

was performed by drop-wise instillation of PBS and recovery

from the same nostril after a few seconds. Bacterial titers in the

nasopharynx and blood, respectively, were determined by serial

dilution, plating, and extrapolation from colony counts the next

day. For determination of bacterial titers in the lungs, mice were

sacrificed, the lungs excised, weighed, and homogenized in PBS.

Bacterial burden in the lungs was assessed per gram of

homogenized tissue.

Figure 1. Differences in surface expression of pneumococcal ligands is dependent on serotype. A) Immunoblot analysis comparingrelative protein levels in whole cell lysates of TIGR4, an unencapsulated derivate (T4R), and the panel of capsular switch mutants. Whole cell lysates(10 mg) were separated by 12%SDS-PAGE and transferred to nitrocellulose or directly blotted onto membranes and probed with antisera to cholinebinding protein (CbpA), pneumococcal surface protein A (PspA), or the pneumococcal serine-rich repeat protein (PsrP). B) Capture of bacteria toimmobilized antibodies to CbpA (black bars) and PsrP (white bars). TIGR4 and the panel of isogenic mutants were incubated in wells pre-coated withserum to CbpA or PsrP and surface expression was indicated by the amount of bacteria captured to immobilized antibody. Values are presented as apercentage of capture of the TIGR4 strain. Statistical analysis was performed using a two-tailed Student’s t-test versus TIGR4. Single asterisk indicatesP,0.05, double asterisks indicates P,0.001.doi:10.1371/journal.pone.0026587.g001



Studies with isolated alveolar macrophages were performed at

University College London under Home Office license PPL70/

6510. Animal experiments were also approved by the University

College London Biological Service Unit Ethics Committee. Sex-

matched 6 week old CD1 mice were inoculated intranasally under

halothane anesthesia with 107 CFU FAM-SE labeled S. pneumoniae.

Mice were sacrificed after 4 hours and bronchoalveolar lavage

fluid (BALF) collected using an angiocatheter and flushing of the

lungs with 1 ml sterile PBS (35). Bacterial CFU were calculated by

plating serial dilutions of BALF. S. pneumoniae phagocytosis by

BALF AMs was analyzed using a FacsCalibur (Becton Dickinson)

to identify the proportion of cells associated with fluorescent

bacteria.

Statistical analysisFor statistical comparisons between 2 cohorts a two-tailed un-

paired Student’s t-test was used. For comparisons between

multiple cohorts a One-Way ANOVA or Dunn’s Multiple

Comparison test was used. Statistical analyses were performed

using SigmaStat software (Systat Software, Chicago IL).

Results

Levels of CbpA, PsrP and PspA were unaffected bychanges in serotype

Prior to experimentation, +6A, +7F, +23F and +4 were

confirmed to produce the appropriate capsule type by multiplex

PCR for serotype specific genes as well as agglutination with

serotype specific antiserum (Figure S1). Additionally, we assessed

whether switching capsule impacted production of proteins. As

determined by immunoblot analysis, +6A, +7F, +23F, had

equivalent total levels of CbpA, which binds to Laminin receptor

and Polymeric immunoglobulin receptor [29,30], PsrP, which

binds to Keratin 10 [13], and PspA, which inhibits complement

Figure 2. Effect of serotype and surface accessibility of pneumococcal ligands. Flow cytometry histograms for pneumococcal surfaceligands CbpA, PspA, and PsrP in A) TIGR4 and its unencapsulated derivative (T4R) and B) the isogenic capsular switch strains +4, +6A, +7F, +23F.Histograms are representative of three independent experiments using bacteria collected from overnight plates (similar results were obtained usingplanktonic pneumococci). Surface accessibility of CbpA, PspA, and PsrP as measured by mean fluorescence intensity (MFI; black bars) and %PE+positively labeled cells (white bars). Results from three independent experiments are shown. Statistical analysis was performed using a two-tailedStudent’s t-test versus T4. Single cross indicates P,0.05, double cross indicates P,0.001.doi:10.1371/journal.pone.0026587.g002



deposition [18,31,39], as +4, TIGR4, and T4R, an unencapsu-

lated TIGR4 derivative (Figure 1A).

Surface accessibility of pneumococcal ligands isdependent on serotype

Using polystyrene plates coated with antibody specific for either

CbpA or PsrP, we first determined that +6A, +7F, and +23F had a

dramatic reduction in their ability to be captured on a flat surface

versus +4. In contrast, +4 was captured at levels equivalent to

TIGR4, and T4R showed enhanced retention (Figure 1B). As the

antibody was immobilized, and therefore unable to penetrate the

capsule layer, this experiment modeled the ability of these adhesins

to bind to their ligand on the host cell surface.

We also compared the ability of free labeled-antibody to bind to

CbpA, PsrP, and PspA on the bacterial surface using flow

cytometry. Compared to TIGR4, T4R had significantly increased

mean fluorescent intensity (MFI) demonstrating that capsule

blocked antibody access to these surface proteins (Figure 2A, C).

Compared to +4, the capsule switch strains +6A, +7F, and +23F

also had reduced accessibility to antibody as determined by MFI

and the detected percentage of PE+ cells (Figure 2B,C). This

remained constant whether bacteria were collected from overnight

blood agar plates (Figure 2) or from logarithmic growth phase

liquid cultures (data not shown). Thus, changes in serotype

reduced surface accessibility of CbpA, PsrP and PspA on the

bacteria outer surface to immobilized and free antibody.

Effect of capsule type on bacterial adhesion in vitroTo assess whether changes in surface accessibility of CbpA and

PsrP altered bacterial adhesion in vitro, we examined the ability of

the capsule switch strains to adhere to Detroit 562 cells, a human

nasopharyngeal cell line, and A549 cells, a type II pneumocyte cell

line (Figure 3). Strain +4, the capsule switch control strain,

adhered to monolayers of both cell lines at levels equivalent to

TIGR4. In contrast, +6A adhered to Detroit 562 and A549 cells at

levels slightly greater and lower than +4, respectively. +7F

demonstrated a moderately enhanced capacity to adhere to these

cells; a 1.5 and 1.7-fold increase in adhesion was observed versus

+4, respectively. Finally, +23F had a less than 2-fold reduction in

its ability to adhere to Detroit 562 cells but not A549 cells. Thus

changes in capsule type resulted in unexceptional but reproducible

changes in bacterial adhesion in vitro that were capsule type and

host cell type dependent.

Capsule type affects nasopharyngeal colonization andvirulence potential

Having observed discrepant results in regards to the impact of

serotype switch on bacterial antibody capture and cell adhesion,

we tested the effect of changing capsule type on nasopharyngeal

colonization and subsequent development of pneumonia; infec-

tious disease states that have previously been shown to require

CbpA and PsrP [37,38]. Importantly, the +4 control strain was

found to be modestly attenuated compared to the TIGR4 strain in

the intransal and intraperitoneal experiments (Figure 4A, 5B).

Thus, insertion of the Janus cassette had inadvertent consequences

and only comparisons between the +6A, +7F, and +23F versus +4

are valid.

Following intranasal challenge, +6A, +7F, and +23F had

significantly reduced median bacterial titers in the nasopharynx

versus +4 as early as 24 h post-infection and continuing up to 120

h (Figure 4A). For +6A a 100-fold reduction in median bacterial

titers was observed after 120 h. Surprisingly, only +4 caused

bacteremia following intranasal challenge (Figure 4B). At no time

were +6A, +7F, or +23F detected in the bloodstream of challenged

mice. Thus the capsule switch strains were moderately attenuated

in their ability to colonize the nasopharynx but completely unable

to cause invasive disease following intranasal challenge.

To dissect how capsule type might be affecting disease

progression, we subsequently infected mice intratracheally and

intraperitoneally (Figure 5). Mice infected intratracheally with

105 CFU of the +6A, +7F, and +23F capsule switch strains had no

bacteria in their lungs or blood at any time tested. In contrast,

mice infected with +4 strain developed low-grade pneumonia and

bacteremia that was comparable to wild type (Figure 5A).

Following intraperitoneal challenge, +6A, +7F were capable of

persisting within the blood at lower levels than +4, whereas +23F

was immediately cleared (Figure 5B). Thus, changes in capsule

type had a strong attenuating affect within the lungs that

precluded entry of +6A, +7F into the bloodstream, and +23F

was attenuated in all anatomical sites tested.

Figure 3. Effect of serotype on bacterial adhesion. In vitro adhesion assays measuring the ability of TIGR4 or the isogenic capsular switch mutantsto adhere to A) human lung alveolar (A549) and B) nasopharyngeal (Detroit) cell lines. Adhesion is expressed as a percentage relative to TIGR4 adhesion.Statistical analysis was performed using a two-tailed Student’s t-test versus TIGR4. Single asterisk indicates P,0.05, double asterisks indicates P,0.001.doi:10.1371/journal.pone.0026587.g003



Capsule switch strains are readily cleared by the hostWithin the lungs, establishment of pneumococcal pneumonia

is dependent on both the ability of the bacteria to attach to cells,

yet remain resistant to phagocytosis by resident alveolar

macrophages. To investigate whether effects on macrophage-

mediated immunity might also explain the strong attenuation of

the capsule switch strains in the lungs, mice were inoculated

with fluorescent S. pneumoniae strains and bacterial uptake by

alveolar macrophages assessed 4 hours after inoculation using

flow cytometry. The association of alveolar macrophages with

+6A and +23F strains was increased compared to the +4 strain,

whereas the association of the +7F strain was similar (Figure 6A).

Bacterial CFU in the BALF was consistent with these results,

with significantly fewer bacterial CFU for the +6A and +23F

but not the +7F compared to +4 in BALF 4 hours after

inoculation (Figure 6B). Hence, serotype 6A and 23F increased

uptake of S. pneumoniae by alveolar macrophages and the rate of

bacterial clearance from the lung, whereas capsule type 7F had

no effect.

Discussion

While considerable work has been done to examine the effect of

capsule type on complement deposition and the effect of serotype

on the ability of pneumococci to colonize the nasopharynx and

survive in bloodstream [4,24,26,27,28,40,41], this is the first study

to assess the impact of capsule type on bacterial adhesin function

and the development of pneumonia. Given that pneumonia is the

most common serious disease presentation caused by the

pneumococcus and host-cell adhesion is a requisite event for

development of invasive disease [15,29,37,38,42], our study begins

to address this important lapse in our understanding of S.

pneumoniae pathogenesis.

The most studied feature of the pneumococcal capsule is its

inhibitory effect on opsonization and phagocytosis. In regards to

the latter, using the same isogenic capsule switch strains, we and

others have shown that capsule type impacts the amount of C3b

that is deposited on the bacteria surface and subsequent uptake by

neutrophils [4,24,25,26,27]. Our current findings suggest that the

Figure 4. Genetic switch of serotype alters nasopharyngeal colonization and virulence in mice. Bacterial titers were measured forindividual 5-week old female BALB/cJ mice intranasally infected with 106 CFU of the parental TIGR4 WT (n = 8) or the isogenic capsular switch strains+6A, +7F, +23F, and +4 (n = 6). A) Nasal lavages and B) blood were collected at 24, 72, and 120 h post-infection. Horizontal bars represent the medianvalue. Statistical analysis was performed using One-Way ANOVA in comparison to both the TIGR4 strain (*) and the internal +4 control strain (+).Asterisk and the cross indicate a significance value of P,0.05.doi:10.1371/journal.pone.0026587.g004



capsule type might also impact other pathogenic mechanisms, in

particular PsrP and CbpA function. In support of this notion, we

observed reduced capture of intact bacteria by immobilized

antibody against PsrP and CbpA in the +6A, +7F, and +23F

capsule switch strains versus +4, altered accessibility of free-

antibody to these adhesins and to PspA by flow cytometry, and a

stronger attenuation for +6A and +7F in the lungs versus the

blood. In the lungs the pneumococcus requires both bacterial

attachment and resistance to opsonophagocytosis in order to

persist. The combination of these two factors may explain the

different effects of serotype on virulence between pneumonia and

sepsis models.

Pneumococcal adhesion is dependent on the ability of bacterial

proteins to interact with ligands on the host cell surface. The

presence of capsule is known to negatively impact these

interactions, thus investigators routinely use unencapsulated

pneumococci when examining the role of their selected adhesins

[42,43,44]. Our in vitro finding that capsule type impacted the

ability of immobilized antibody against CbpA and PsrP to capture

pneumococci suggests that the biochemical properties of these

capsule types alters their exposure on the outer surface of the

pneumococcus. Results from our cytometric analyses support this

notion and showed that the presence of capsule decreased the

ability of free antibody against PspA, CbpA, and PsrP to gain

access to these proteins, and that capsule types 6A, 7F, and 23F

permitted less accessibility that capsule type 4. Notably, PsrP and

CbpA deficient mutants have been demonstrated to be more

severely attenuated in the lungs rather than bloodstream of

experimentally infected mice [37,38]. Thus, the observed impact

of diminished CbpA and PsrP exposure by +6A, +7F, and +23F

versus +4, is consistent with past and present experimental

observations in vivo.

Importantly, the modest reduction observed for only +23F in

the in vitro bacterial adhesion assays is difficult to reconcile with our

data showing reduced surface accessibility of CbpA and PsrP. This

discrepancy may be because the pneumococcus uses multiple

Figure 5. Serotype affects virulence in the lungs. Bacterial titers were measured for individual 5-week old female BALB/cJ mice infected via A)intratracheal (105 CFU) or B) intraperitoneal (104 CFU) routes with TIGR4 WT (n = 6) or the isogenic capsular switch strains +6A, +7F, +23F, and +4(n = 6). For intratracheal challenge samples were collected at 48 h post-infection, and for intraperitoneal challenge samples were collected at 24 and48 hours. Horizontal bars represent the median value. Statistical analysis was performed using One-Way ANOVA in comparison to both the TIGR4strain (*) and the internal +4 control strain (+). Asterisk and the cross indicate a significance value of P,0.05.doi:10.1371/journal.pone.0026587.g005



adhesins to attach to host-cells and other unaffected adhesins

compensated. For example, the pneumococcus also binds to

epithelial cells through phosphorylcholine residues on its cell wall,

with the tip of a pilus structure, using the MSCRAMMs (Microbial

Surface Components Recognizing Adhesive Matrix Molecules)

PavA and PavB, and with a multitude of normally cytoplasmic

enzymes that decorate the bacterial surface, among others

[45,46,47]. If this compensatory effect were to hold true in vivo,

the observed dramatic attenuation of the isogenic switch strains in

the lungs would therefore have to be the result of capsule-mediated

changes in the secondary function of surface proteins. While we

did not explore this possibility, CbpA has been shown to bind to

Factor H and further inhibit complement deposition [48].

Likewise, PsrP also mediates in vivo biofilm formation [49]. Thus

altered function on top of reduced accessibility may explain the

discrepancy between cell adhesion assays and virulence.

CbpA, PspA, and PsrP are highly variable and show strong

associations with clonotype and/or capsular type. For example

numerous isotypes of CbpA exists with alterations in the R1 and

R2 domains of the protein [50]. Similarly, the protein PspA is

classified into two major families depending on its amino acid

composition [31]. Sequence analysis of the published genomes

indicates that PsrP also demonstrates considerable variability, in

particular the length of its SRR2 domain that serves to extend the

protein outward [13]. The observed difference in the surface

accessibility of these proteins and their known structural variability

raises the possibility that some versions of CbpA and PsrP might

only be compatible with certain serotypes. Perhaps invasive clones

of each serotype carry the optimized version of these proteins for

the corresponding serotype. This would perhaps help explain why

certain virulence determinants such as PsrP are not equally

distributed among serotypes [17]. Interestingly, in the MFI

experiments with the capsule switch strains and free antibody

(Figure 2B), we observed that some of the graphs had what might

be construed as two distinct cell populations (e.g. PspA). As we

used confirmed opaque bacteria, we could exclude phase-variation

as the cause of this bimodality. Furthermore, as we saw similar

results following experiments with overnight cultures and plank-

tonic cultures, we could exclude the presence of dead cells. The

cause for this phenomena remains unknown, albeit for the

pneumococcal pilus a non-phase variable bimodilty has been

recently described [51].

Finally, we have previously demonstrated that complement

activity was maximum against +6A and +23F, followed by +7F,

and then +4 [24]; and this was reflected in our current

experiments examining bacterial uptake by resident alveolar

macrophages in the lungs, measurement of bacterial titers in the

bronchoalveolar lavage fluid, and after intraperitoneal inoculation

where mice challenged with +6A, +7F, and +23F had significantly

lower numbers of pneumococci in their blood than those

challenged with +4. Thus any effects of capsule on CbpA and

PsrP function are concomitant with those that occur in regards to

opsonophagocytosis.

Importantly, all the strains used in this study were of the

opaque phase variation, had similar capsule thickness, and

showed equivalent growth rates in vitro [24]. While we were

surprised to observe that the +4 strain was attenuated in vivo in

comparison to wild type TIGR4, suggesting an inadvertent affect

of the Janus insertion or another mutation incurred during

creation, the fact that we only compared +6A, +7F, +23F to +4 in

these studies allowed for its control. Thus the observed differences

can be attributed to the biological properties of the serotype and

not the relative fitness or amount of capsule surrounding the

bacterium.

Taken together our findings indicate that there is complex

relationship between capsular and non-capsular determinants that

synergistically effect virulence potential, particularly in the lungs.

Based on our findings and those of other investigators, we propose

that the capsular type imposes a biochemical threshold for surface

accessibility/functionality of proteins which must be overcome by

altered or enhanced expression of other genetic determinants. As a

result, certain genetic backgrounds (i.e. clonotypes) are better

suited for certain serotypes versus others. We speculate that this

explains why certain forms of proteins are expressed in specific

capsular type backgrounds, moreover, why we observed that

replacement of capsular type can have a profound effect on

virulence independent of strain background.

Supporting Information

Figure S1 Confirmation of expression of capsule typesby the capsular switch strains. A) Appropriate capsular

serotype production was confirmed for the switch strains by a

series of five different PCR reactions which amplified serotype-

specific genes for capsule type 4, 6A, 7F, and 23F as well as cpsA

(positive control). B) Agglutination assay using serotype specific

antiserum confirmed production of capsule types by the capsule

switch mutants.

(EPS)

Figure 6. Clearance of FAMSE-labeled capsule switch strainsfrom the lungs and uptake by alveolar macrophages 4 hours-post inoculation. A) Alveolar macrophage phagocytosis of CPSswitch strains as determined by FACS. Data presented as the median(IQR) fluorescence index (percentage of fluorescent macrophagesmultiplied by the geo mean MFI) for BALF macrophages (n = 5). B)Median bacterial titers in bronchoalveolar lavage fluid (BALF) (n = 5). Forpanels A and B samples were collected 4 hours after intranasalchallenge with 56106 CFU FAMSE-labeled CFU, and the P values givenabove the respective box and whisker plot for individual strains are forcomparison to the +4 strain. A Dunn’s Multiple Comparison test wasused for statistical analyses.doi:10.1371/journal.pone.0026587.g006



Acknowledgments

We would like to thank Dr. Jeffery Weiser at the University of Pennsylvania

School of Medicine and Dr. Elaine Tuomanen at St. Jude Children’s

Research Hospital for the gifts of the capsular switch strains and antibodies

against CbpA, respectively. We also thank Carla M. Gorena and Benjamin

J. Daniel at the University of Texas Health Science Center Flow

Cytometry Core Facility for their assistance in processing samples for

analyses.

Author Contributions

Conceived and designed the experiments: CJS JSB CJO. Performed the

experiments: CJS CH PS CAH CJO EC. Analyzed the data: CJS CAH

JSB CJO. Contributed reagents/materials/analysis tools: JSB CJO. Wrote

the paper: CJS JSB CJO.

References

1. Kadioglu A, Weiser JN, Paton JC, Andrew PW (2008) The role of Streptococcus

pneumoniae virulence factors in host respiratory colonization and disease. Nat Rev

Microbiol 6: 288–301.

2. Wartha F, Beiter K, Albiger B, Fernebro J, Zychlinsky A, et al. (2007) Capsule

and D-alanylated lipoteichoic acids protect Streptococcus pneumoniae against

neutrophil extracellular traps. Cell Microbiol 9: 1162–1171.

3. Kim JO, Romero-Steiner S, Sorensen UB, Blom J, Carvalho M, et al. (1999)

Relationship between cell surface carbohydrates and intrastrain variation on

opsonophagocytosis of Streptococcus pneumoniae. Infect Immun 67: 2327–2333.

4. Hyams C, Camberlein E, Cohen JM, Bax K, Brown JS (2010) The Streptococcus

pneumoniae capsule inhibits complement activity and neutrophil phagocytosis by

multiple mechanisms. Infect Immun 78: 704–715.

5. Briles DE, Crain MJ, Gray BM, Forman C, Yother J (1992) Strong association

between capsular type and virulence for mice among human isolates of

Streptococcus pneumoniae. Infect Immun 60: 111–116.

6. Morona JK, Miller DC, Morona R, Paton JC (2004) The effect that

mutations in the conserved capsular polysaccharide biosynthesis genes cpsA,

cpsB, and cpsD have on virulence of Streptococcus pneumoniae. J Infect Dis 189:

1905–1913.

7. Morona JK, Morona R, Paton JC (2006) Attachment of capsular polysaccharide

to the cell wall of Streptococcus pneumoniae type 2 is required for invasive

disease. Proc Natl Acad Sci U S A. 103: 8505–8510.

8. Lu PJ, Nuorti JP (2010) Pneumococcal polysaccharide vaccination among adults

aged 65 years and older, U.S., 1989-2008. Am J Prev Med 39: 287–295.

9. Nuorti JP, Whitney CG (2010) Prevention of pneumococcal disease among

infants and children - use of 13-valent pneumococcal conjugate vaccine and 23-

valent pneumococcal polysaccharide vaccine - recommendations of the Advisory

Committee on Immunization Practices (ACIP). MMWR Recomm Rep 59:

1–18.

10. Talbot UM, Paton AW, Paton JC (1996) Uptake of Streptococcus pneumoniae by

respiratory epithelial cells. Infect Immun 64: 3772–3777.

11. Hammerschmidt S, Wolff S, Hocke A, Rosseau S, Muller E, et al. (2005)

Illustration of pneumococcal polysaccharide capsule during adherence and

invasion of epithelial cells. Infect Immun 73: 4653–4667.

12. Kim JO, Weiser JN (1998) Association of intrastrain phase variation in quantity

of capsular polysaccharide and teichoic acid with the virulence of Streptococcus

pneumoniae. J Infect Dis 177: 368–377.

13. Shivshankar P, Sanchez C, Rose LF, Orihuela CJ (2009) The Streptococcus

pneumoniae adhesin PsrP binds to Keratin 10 on lung cells. Mol Microbiol 73:

663–679.

14. Hilleringmann M, Ringler P, Muller SA, De Angelis G, Rappuoli R, et al. (2009)

Molecular architecture of Streptococcus pneumoniae TIGR4 pili. EMBO J 28:

3921–3930.

15. Nelson AL, Ries J, Bagnoli F, Dahlberg S, Falker S, et al. (2007) RrgA is a pilus-

associated adhesin in Streptococcus pneumoniae. Mol Microbiol 66: 329–340.

16. Imai S, Ito Y, Ishida T, Hirai T, Ito I, et al. (2011) Distribution and clonal

relationship of cell surface virulence genes among Streptococcus pneumoniae isolates

in Japan. Clin Microbiol Infect 17: 1409–1414.

17. Munoz-Almagro C, Selva L, Sanchez CJ, Esteva C, de Sevilla MF, et al. (2010)

PsrP, a protective pneumococcal antigen, is highly prevalent in children with

pneumonia and is strongly associated with clonal type. Clin Vaccine Immunol

17: 1672–1678.

18. Quin LR, Moore QC 3rd, McDaniel LS (2007) Pneumolysin, PspA, and PspC

contribute to pneumococcal evasion of early innate immune responses during

bacteremia in mice. Infect Immun 75: 2067–2070.

19. World Health Organization (WHO) (1999) Pneumococcal vaccines. WHO

position paper. Wkly Epidemiol Rec 74: 177–183.

20. Lexau CA, Lynfield R, Danila R, Pilishvili T, Facklam R, et al. (2005) Changing

epidemiology of invasive pneumococcal disease among older adults in the era of

pediatric pneumococcal conjugate vaccine. JAMA 294: 2043–2051.

21. Obert C, Sublett J, Kaushal D, Hinojosa E, Barton T, et al. (2006)

Identification of a candidate Streptococcus pneumoniae core genome and regions

of diversity correlated with invasive pneumococcal disease. Infect Immun 74:

4766–4777.

22. Sandgren A, Sjostrom K, Olsson-Liljequist B, Christensson B, Samuelsson A,

et al. (2004) Effect of clonal and serotype-specific properties on the invasive

capacity of Streptococcus pneumoniae. J Infect Dis 189: 785–796.

23. Kelly T, Dillard JP, Yother J (1994) Effect of genetic switching of capsular type

on virulence of Streptococcus pneumoniae. Infect Immun 62: 1813–1819.

24. Hyams C, Yuste J, Bax K, Camberlein E, Weiser JN, et al. (2010) Streptococcus

pneumoniae resistance to complement-mediated immunity is dependent on the

capsular serotype. Infect Immun 78: 716–725.

25. Melin M, Jarva H, Siira L, Meri S, Kayhty H, et al. (2009) Streptococcus pneumoniae

capsular serotype 19F is more resistant to C3 deposition and less sensitive to

opsonophagocytosis than serotype 6B. Infect Immun 77: 676–684.

26. Melin M, Trzcinski K, Antonio M, Meri S, Adegbola R, et al. (2010) Serotype-

related variation in susceptibility to complement deposition and opsonophago-cytosis among clinical isolates of Streptococcus pneumoniae. Infect Immun 78:

5252–5261.

27. Melin M, Trzcinski K, Meri S, Kayhty H, Vakevainen M (2010) The capsularserotype of Streptococcus pneumoniae is more important than the genetic background

for resistance to complement. Infect Immun 78: 5262–5270.

28. Yuste J, Khandavilli S, Ansari N, Muttardi K, Ismail L, et al. (2010) The

effects of PspC on complement-mediated immunity to Streptococcus pneumoniae

vary with strain background and capsular serotype. Infect Immun 78:

283–292.

29. Zhang JR, Mostov KE, Lamm ME, Nanno M, Shimida S, et al. (2000) Thepolymeric immunoglobulin receptor translocates pneumococci across human

nasopharyngeal epithelial cells. Cell 102: 827–837.

30. Orihuela CJ, Mahdavi J, Thornton J, Mann B, Wooldridge KG, et al. (2009)

Laminin receptor initiates bacterial contact with the blood brain barrier inexperimental meningitis models. J Clin Invest 119: 1638–1646.

31. Ren B, Szalai AJ, Thomas O, Hollingshead SK, Briles DE (2003) Both family 1

and family 2 PspA proteins can inhibit complement deposition and confervirulence to a capsular serotype 3 strain of Streptococcus pneumoniae. Infect Immun

71: 75–85.

32. Abeyta M, Hardy GG, Yother J (2003) Genetic alteration of capsule type but not

PspA type affects accessibility of surface-bound complement and surface antigensof Streptococcus pneumoniae. Infect Immun 71: 218–225.

33. Tettelin H, Nelson KE, Paulsen IT, Eisen JA, Read TD, et al. (2001) Complete

genome sequence of a virulent isolate of Streptococcus pneumoniae. Science 293:

498–506.

34. Mann B, Orihuela C, Antikainen J, Gao G, Sublett J, et al. (2006)Multifunctional role of choline binding protein G in pneumococcal pathogen-

esis. Infect Immun 74: 821–829.

35. Brito DA, Ramirez M, de Lencastre H (2003) Serotyping Streptococcus pneumoniae

by multiplex PCR. J Clin Microbiol 41: 2378–2384.

36. Pai R, Gertz RE, Beall B (2006) Sequential multiplex PCR approach for

determining capsular serotypes of Streptococcus pneumoniae isolates. J Clin

Microbiol 44: 124–131.

37. Rose L, Shivshankar P, Hinojosa E, Rodriguez A, Sanchez CJ, et al. (2008)Antibodies against PsrP, a novel Streptococcus pneumoniae adhesin, block adhesion

and protect mice against pneumococcal challenge. J Infect Dis 198: 375–383.

38. Orihuela CJ, Gao G, Francis KP, Yu J, Tuomanen EI (2004) Tissue-specific

contributions of pneumococcal virulence factors to pathogenesis. J Infect Dis190: 1661–1669.

39. Yuste J, Botto M, Paton JC, Holden DW, Brown JS (2005) Additive inhibition of

complement deposition by pneumolysin and PspA facilitates Streptococcus

pneumoniae septicemia. J Immunol 175: 1813–1819.

40. Lysenko ES, Lijek RS, Brown SP, Weiser JN (2010) Within-host competition

drives selection for the capsule virulence determinant of Streptococcus pneumoniae.

Curr Biol 20: 1222–1226.

41. Sleeman KL, Griffiths D, Shackley F, Diggle L, Gupta S, et al. (2006) Capsularserotype-specific attack rates and duration of carriage of Streptococcus pneumoniae in

a population of children. J Infect Dis 194: 682–688.

42. Holmes AR, McNab R, Millsap KW, Rohde M, Hammerschmidt S, et al.(2001) The pavA gene of Streptococcus pneumoniae encodes a fibronectin-binding

protein that is essential for virulence. Mol Microbiol 41: 1395–1408.

43. Munoz-Elias EJ, Marcano J, Camilli A (2008) Isolation of Streptococcus pneumoniae

biofilm mutants and their characterization during nasopharyngeal colonization.Infect Immun 76: 5049–5061.

44. Daniely D, Portnoi M, Shagan M, Porgador A, Givon-Lavi N, et al. (2006)

Pneumococcal 6-phosphogluconate-dehydrogenase, a putative adhesin, induces

protective immune response in mice. Clin Exp Immunol 144: 254–263.

45. Bergmann S, Hammerschmidt S (2006) Versatility of pneumococcal surfaceproteins. Microbiology 152: 295–303.

46. Kline KA, Falker S, Dahlberg S, Normark S, Henriques-Normark B (2009)

Bacterial adhesins in host-microbe interactions. Cell Host Microbe 5:580–592.



47. Paterson GK, Orihuela CJ (2010) Pneumococcal microbial surface components

recognizing adhesive matrix molecules targeting of the extracellular matrix. MolMicrobiol 77: 1–5.

48. Dave S, Brooks-Walter A, Pangburn MK, McDaniel LS (2001) PspC, a

pneumococcal surface protein, binds human factor H. Infect Immun 69:3435–3437.

49. Sanchez CJ, Shivshankar P, Stol K, Trakhtenbroit S, Sullam PM, et al. (2010)The pneumococcal serine-rich repeat protein is an intra-species bacterial

adhesin that promotes bacterial aggregation in vivo and in biofilms. PLoS

Pathog 6: e1001044.50. Iannelli F, Chiavolini D, Ricci S, Oggioni MR, Pozzi G (2004) Pneumococcal

surface protein C contributes to sepsis caused by Streptococcus pneumoniae in mice.

Infect Immun 72: 3077–3080.51. De Angelis G, Moschioni M, Muzzi A, Pezzicoli A, Censini S, et al. (2011) The

Streptococcus pneumoniae pilus-1 displays a biphasic expression pattern. PloS One 6:e21269.



Changes in Capsular Serotype Alter the Surface Exposure of Pneumococcal Adhesins and Impact...

Documents

Transcript of Changes in Capsular Serotype Alter the Surface Exposure of Pneumococcal Adhesins and Impact...